Multichannel Digital Transmission in an Optical Network of

In present telecommunication networks, information transfer relies on the interplay of optical and electrical signals. Data are communicated optically...
3 downloads 0 Views 323KB Size
Published on Web 02/09/2002

Multichannel Digital Transmission in an Optical Network of Communicating Molecules Franc¸ isco M. Raymo* and Silvia Giordani Contribution from the Center for Supramolecular Science, Department of Chemistry, UniVersity of Miami, 1301 Memorial DriVe, Coral Gables, Florida 33146-0431 Received August 23, 2001

Abstract: In present telecommunication networks, information transfer relies on the interplay of optical and electrical signals. Data are communicated optically but processed electronically. Methods to maintain the propagating signals solely at the optical level must be developed to overcome the transmission capacities and speed limits imposed by the electronic components. We have demonstrated that molecular switches can be used to gate optical signals in response to optical signals. We have realized a simple optical network consisting of three light sources, one cell containing a solution of three fluorescent molecules, one cell containing a solution of a three-state molecular switch and a detector. The light emitted by the three fluorophores is absorbed by the three states of the molecular switch. Using this simple operating principle, we have shown that multichannel digital transmission can be implemented on an ensemble of communicating molecules relying exclusively on the interplay of optical inputs and optical outputs.

Introduction

The growing demand for telecommunication and Internet applications continues to stimulate the development of optical networks for the faster transmission of larger volumes of data.1 In these networks, bundles of optical fibers coupled to optoelectronic devices ensure the communication of optical signals over long distances. Present optical transport technology permits the transmission of hundreds of gigabits per second over hundreds of kilometers. The routes of the optical signals traveling through these networks are determined by switching devices in response to electrical stimulations. Unfortunately, the interplay of optical and electrical signals in these hybrid devices limits dramatically the transmission capacity and speed.1,2 Only a tiny fraction of the huge bandwidth of optical fibers can be used. In principle, hundreds of optical signals with closely spaced wavelengths can be transmitted through a single waveguide relying on current optical fiber technology.1c The propagating light beams do not interfere with each other and can be transported in parallel. The noninteracting properties of optical signals, however, do not apply to electrical signals. Only one electrical signal can be transported through a single electrical wire. Thus, even although optical fibers can transmit multiple data in parallel, optoelectronic switches process them more or less sequentially.1b The electronic portion of these devices simply cannot handle the immense parallelism potentially offered by optical signals. These intrinsic limitations will stop (1) (a) Special issue on Optical Networking. Bell Labs Technol. J. 1999, 4 (1), 3-322. (b) Franz, J. H.; Jain, V. K. Optical Communications: Components and Systems; CRC Press: Boca Raton, FL, 2000. (c) Mynbaev, D. K.; Scheiner, L. L. Fiber Optic Communications Technology; Prentice Hall: Upper Saddle River, NJ, 2001. (2) (a) Kahn, J. M.; Ho, K.-P. Nature 2001, 411, 1007-1010. (b) Mitra, P. P.; Stark, J. B. Nature 2001, 411, 1027-1030. 2004 VOL. 124, NO. 9, 2002

9

J. AM. CHEM. SOC.

soon the rapid escalation of the photonic traffic supported in optical networks.2 It is necessary to develop practical strategies to switch the propagating optical signals with optical, rather than electrical, stimulations.3 Ultimately, the electronic element of present communication networks must be eliminated completely, all intervening signals must be optical and only optical effects must be exploited. Molecular switches4 are promising candidates for the realization of signal processing networks. Simple logic operations5 have been implemented already at the molecular level by using chemical, electrical, and/or optical signals.6-12 Furthermore, the miniaturized dimensions of these chemical systems have encouraged the design of prototypical devices incorporating (3) (a) Thylen, L.; Karlsson, G.; Nilsson, O. IEEE Commun. Mag. 1996, 34 (2), 106-113. (b) Nolte, D. D. J. Appl. Phys. 1999, 85, 6259-6289. (c) Jackman, N. A.; Patel, S. H.; Mikkelsen, B. P.; Korotky, S. K. Bell Labs Technol. J. 1999, 4 (1), 262-281. (d) McCarthy, D. C. Photonics Spectra 2001, 35 (3), 140-150. (e) Veeraraghavan, M.; Karri, R.; Moors, T.; Karol, M.; Grobler, R. IEEE Commun. Magn. 2001, 39 (3), 118-127. (4) (a) Special issue on Photochromism: Memories and Switches. Chem. ReV. 2000, 100, 1683-1890. (b) Molecular Switches; Feringa, B. L., Ed.; WileyVCH: Weinheim, Germany, 2001. (5) Mitchell, R. J. Microprocessor Systems: An Introduction; Macmillan: Houndsmill, UK, 1995. (6) (a) de Silva, A. P.; Gunaratne, H. Q. N.; McCoy, C. P. Nature 1993, 364, 42-44. (b) de Silva, A. P.; Gunaratne, H. Q. N.; McCoy, C. P. J. Am. Chem. Soc. 1997, 119, 7891-7892. (c) de Silva, A. P.; Dixon, I. M.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Maxwell, P. R.S.; Rice, T. E. J. Am. Chem. Soc. 1999, 121, 1393-1394. (d) de Silva, A. P.; McClenaghan, N. D. J. Am. Chem. Soc. 2000, 122, 3965-3966. (7) (a) Asakawa, M.; Ashton, P. R.; Balzani, V.; Credi, A.; Mattersteig, G.; Matthews, O. A.; Montalti, M.; Spencer, N.; Stoddart, J. F.; Venturi, M. Chem. Eur. J. 1997, 3, 1992-1996. (b) Credi, A.; Balzani, V.; Langford, S. J.; Stoddart, J. F. J. Am. Chem. Soc. 1997, 119, 2679-2681. (c) Collier, C. P.; Wong, E. W.; Belohradsky, M.; Raymo, F. M.; Stoddart, J. F.; Kuekes, P. J.; Williams, R. S.; Heath, J. R. Science 1999, 285, 391-394. (8) (a) Pina, F.; Roque, A.; Melo, M. J.; Maestri, I.; Belladelli, L.; Balzani, V. Chem. Eur. J. 1998, 4, 1184-1191. (b) Pina, F.; Maestri, M.; Balzani, V. Chem. Commun. 1999, 107-114. (c) Roque, A.; Pina, F.; Alves, S.; Ballardini, R.; Maestri, M.; Balzani, V. J. Mater. Chem. 1999, 9, 22652269. (d) Pina, F.; Melo, M. J.; Maestri, M.; Passaniti, P.; Balzani, V. J. Am. Chem Soc. 2000, 122, 4496-4498. 10.1021/ja016920e CCC: $22.00 © 2002 American Chemical Society

Multichannel Digital Transmission

ARTICLES

Figure 2. The optical network employed to implement digital transmission on an ensemble of communicating molecules.

associated with these chemical transformations can be exploited to switch optical signals. Results and Discussion

Figure 1. The switching cycle associated with the three states SP, ME, and MEH.

molecular components.13,14 However, the lack of reliable strategies to communicate signals between individual molecules has prevented, so far, the integration of molecular switches into functioning circuits.14e Recently, we have developed a threestate molecular switch15 and we have identified a simple mechanism for the intermolecular communication of optical signals.16 The design of our switch is based on the reversible photoisomerization of a spiropyran derivative.17 Ultraviolet light, visible light, and H+ inputs induce the interconversion between the colorless spiropyran state SP (Figure 1) and the colored merocyanine forms ME and MEH. In this article, we demonstrate that the pronounced changes in absorption properties (9) (a) Gobbi, L.; Seiler, P.; Diederich, F. Angew. Chem., Int. Ed. 1999, 38, 674-678. (b) Gobbi, L.; Seiler, P.; Diederich, F.; Gramlich, V. HelV. Chim. Acta 2000, 83, 1711-1723. (c) Gobbi, L.; Seiler, P.; Diederich, F.; Gramlich, V.; Boudon, C.; Gisselbrecht, J. P.; Gross, M. HelV. Chim. Acta 2001, 84, 743-777. (10) (a) Debreczeny, M. P.; Svec, W. A.; Wasielewski, M. R. Science 1996, 274, 584-587. (b) Gosztola, D.; Niemczyk, M. P.; Wasielewski, M. R. J. Am. Chem. Soc. 1998, 120, 5118-5119. (c) Hayes, R. T.; Wasielewski, M. R.; Gosztola, D. J. Am. Chem. Soc. 2000, 122, 5563-5567. (d) Lukas, A. S.; Bushard, P. J.; Wasielewski, M. R. J. Am. Chem. Soc. 2001, 123, 2440-2441. (11) (a) Parker, D.; Williams, J. A. G. Chem. Commun. 1998, 245-246. (b) Gunnlaugsson, T.; MacDonail, D. A.; Parker, D. Chem. Commun. 2000, 93-94. (12) Ji, H. F.; Dabestani, R.; Brown, G. M. J. Am. Chem. Soc. 2000, 122, 93069307. (13) (a) Bard, A. J. Integrated Chemical Systems: A Chemical Approach to Nanotechnology; Wiley: New York, 1994. (b) Special issue on Molecular Machines. Acc. Chem. Res. 2001, 34, 409-522. (14) (a) Goldhaber-Gordon, D.; Montemerlo, M. S.; Love, J. C.; Opiteck, G. J.; Ellenbogen, J. C. Proc. IEEE 1997, 85, 521-540. (b) Metzger, R. M. Acc. Chem. Res. 1999, 32, 950-957. (c) Tour, J. M. Acc. Chem. Res. 2000, 33, 791-804. (d) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. Angew. Chem., Int. Ed. 2000, 39, 3348-3391. (e) Joachim, C.; Gimzewski, J. K.; Aviram, A. Nature 2000, 408, 541-548. (f) Heath, J. R. Pure Appl. Chem. 2000, 72, 11-20. (15) Raymo, F. M.; Giordani, S. J. Am. Chem. Soc. 2001, 123, 4651-4652. (16) Raymo, F. M.; Giordani, S. Org. Lett. 2001, 3, 1833-1836. (17) (a) Bertelson, R. C. Photochrosmism; Brown, G. H., Ed.; Wiley: New York, 1971; pp 45-431. (b) Guglielmetti, R. Photochromism: Molecules and Systems; Du¨rr, H., Bouas-Laurent, H., Eds.; Elsevier: Amsterdam, The Netherlands, 1990; pp 314-466 and 855-878. (c) Willner, I.; Willner, B. Bioorganic Photochemistry; Morrison, H., Ed.; Wiley: New York, 1993; pp 1-110. (d) Willner, I.; Rubin, S. Angew. Chem., Int. Ed. Engl. 1996, 35, 367-385. (e) Willner, I. Acc. Chem. Res. 1997, 30, 347-356. (f) Bertelson, R. C. Organic Photochromic and Thermochromic Compounds; Crano, J. C., Guglielmetti, R., Eds.; Plenum Press: New York, 1999; pp 11-83. (g) Willner, I.; Katz, E. Angew. Chem., Int. Ed. 2000, 39, 11801218. (h) Berkovic, G.; Krongauz, V.; Weiss, V. Chem. ReV. 2000, 100, 1741-1753. (i) Shipway, A. N.; Willner, I. Acc. Chem. Res. 2001, 34, 421-432.

The simple optical network shown in Figure 2 was assembled in the sample compartment of a commercial emission spectrometer. An optical signal (yellow arrows) travels from the excitation source to the detector passing through two quartz cells. Cell A contains a solution of fluorescent molecules that absorb the incident radiation and reemit light at longer wavelengths. The light emitted in the direction perpendicular to the exciting beam is communicated to cell B. This cell contains a solution of the three-state molecular switch illustrated in Figure 1. The interconversion between SP, ME, and MEH is induced by addressing cell B with three input signals (red arrows). They are ultraviolet light (I1), visible light (I2), and H+ (I3). These inputs modulate the amount of light absorbed by the solution in cell B and, ultimately, determine the intensity of the final optical output (O1) that reaches the detector. The fluorescent molecules selected for this investigation are naphthalene (NA), anthracene (AN), and tetracene (TE). Their emission spectra (a, b, and c in Figure 3) were recorded by placing a MeCN solution of the fluorophore in cell A and pure MeCN in cell B. The absorption spectra (d, e, and f in Figure 3) of the three states of the molecular switch are significantly different in the range of wavelengths where NA, AN, and TE emit. The absorbances of SP, ME, and MEH at wavelengths corresponding to selected emission maxima of NA, AN, and TE are compared in the pie charts of Figure 3. The NA emission at 335 nm can be absorbed equally by SP, ME, and MEH (g, h, and i in Figure 3). The AN emission at 401 nm can be absorbed only by ME and MEH (j and k in Figure 3). The TE emission at 544 nm can be absorbed only by ME (l in Figure 3). These observations anticipate that (1) NA can communicate with all three states of the molecular switch, (2) AN can communicate with ME and MEH, but not with SP, and (3) TE can communicate with ME, but not with SP and MEH. The intermolecular communication between the three fluorophores and the three states of the molecular switch was explored by filling cell A with an equimolar MeCN solution of NA, AN, and TE and cell B with a MeCN solution of SP. The three fluorophores were excited selectively by irradiating cell A with three consecutive monochromatic beams at 275, 357, and 441 nm. These sequential stimulations induce the emission of NA at 335 nm (NA channel), AN at 401 nm (AN channel), and TE at 544 nm (TE channel), respectively. After each excitation step, the intensity of the optical output O1 was recorded at the wavelength of the stimulated channel. Then, J. AM. CHEM. SOC.

9

VOL. 124, NO. 9, 2002 2005

ARTICLES

Raymo and Giordani

Figure 3. Emission spectra of (a) NA (λexc. ) 275 nm), (b) AN (λexc. ) 357 nm), and (c) TE (λexc. ) 441 nm) recorded using the configuration illustrated in Figure 2 by placing the fluorescent solution (1 × 10-5 M, MeCN) in cell A and pure MeCN in cell B. Absorption spectra of SP (1 × 10-4 M, MeCN) (d) before and (e) after irradiation with ultraviolet light and (f) after irradiation with ultraviolet light followed by the addition of 1 equiv of CF3CO2H. Absorbance of (g) SP, (h) ME, and (i) MEH at 335 nm, of (j) ME and (k) MEH at 401 nm, and of (l) ME at 544 nm.

the interconversion of the molecular switch was induced addressing cell B with the input signals I1, I2, and/or I3 and the intensity of O1 for the three channels was recorded again in three consecutive runs. Following this experimental protocol, the changes in the relative intensities of the three output channels were monitored for nine consecutive switching steps (a-j in Figure 4). Ultraviolet light (I1) and visible light (I2) inputs induce the switching between SP and ME (Figure 1) in cell B. These two states absorb equally at 335 nm (g and h in Figure 3). As a result, their interconversion does not affect the intensity of O1 for the NA channel (a-h in Figure 4 top). The colorless state SP does not absorb at 401 and 544 nm, while the colored form ME absorbs strongly at these wavelengths (j and l in Figure 3). Consistently, the switching from SP to ME, induced by the input I1, reduces the intensities of O1 for the AN and TE channels (a and b in Figure 4 center and bottom). The original values of O1 are restored for both channels (b and c in Figure 4 center and bottom) after the switching from ME back to SP induced by the input of I2. Thus, the alternation of the optical inputs I1 and I2 switches the optical output O1 for the AN and TE channels between two values. The data points a-g in Figure 4 (center and bottom) illustrate this effect for three consecutive switching cycles. The subsequent irradiation with ultraviolet light (I1) produces again ME and the associated decrease in the intensity of O1 for the AN and TE channels. At this point, the addition of 1 equiv of CF3CO2H (I3) to cell B converts ME into MEH (Figure 1). The resulting state MEH absorbs at 2006 J. AM. CHEM. SOC.

9

VOL. 124, NO. 9, 2002

Figure 4. Changes in the relative intensities of O1 for the NA, AN, and TE channels. The intensities of O1 are reported relative to those of the emission bands in a, b, and c of Figure 3. They were recorded with the configuration illustrated in Figure 2 by placing an equimolar solution of NA, AN, and TE (1 × 10-5 M, MeCN) in cell A and a solution of SP (1 × 10-4 M, MeCN) in cell B. Cell A was excited at 275, 357, and 441 nm in three consecutive runs. Cell B was addressed with ultraviolet light (I1), visible light (I2), and H+ (I3).

401 nm (k in Figure 3), but not at 544 nm. As a result, the intensity of O1 for the AN channel remains constant (h and i in Figure 4 center), while that for the TE channel returns to the original value (h and i in Figure 4 bottom). The subsequent irradiation of cell B with visible light (I2) induces the interconversion of MEH into SP (Figure 1). The result is a change in the absorbance of cell B at 401 nm, but not at 544 nm. Thus, the intensity of the AN channel returns to its initial value (i and j in Figure 4 center) while that of the TE channel remains constant (i and j in Figure 4 bottom). In the simple optical network of Figure 2, the three input signals I1, I2, and I3 (red arrows) modulate the optical output O1 (yellow arrows). This switching protocol can be described with the aid of binary logic.5 The three inputs I1, I2, and I3 can be either off or on and each of them can be represented by a binary digit (0 or 1). Similarly, the output O1 can be considered off when its relative intensity is below 50% and on when it is above. Thus, the communicating ensemble of molecules converts an input string of four digits (I1, I2, and I3) into a single output data (O1) for each channel (Table 1). For example, the input string is 010 when the ultraviolet light source is off (I1 ) 0), the visible light source is on (I2 ) 1), and the H+ input is off (I3 ) 0). Under these conditions, the molecular switch is in state SP. This state absorbs the emission of NA at 335 nm (NA channel ) off), but not that of AN at 401 nm (AN channel ) on) and that of TE at 401 nm (TE channel ) on). As a result, the output data O1 corresponding to the input string 010 is 0 for the NA channel, 1 for the AN channel, and 1 for the TE channel. A similar analysis can be extended to all of the eight possible combinations of input strings (Table 1). In the case of the NA channel, the output O1 is always

Multichannel Digital Transmission

ARTICLES

Table 1. Truth Table for the NA, AN, and TE Channels input dataa

output dataa

ultraviolet light I1

visible light I2

H+ I3

NA channelb O1

AN channelc O1

TE channeld O1

0 0 0 1 0 1 1 1

0 0 1 0 1 0 1 1

0 1 0 0 1 1 0 1

0 0 0 0 0 0 0 0

1 0 1 0 1 0 0 0

1 1 1 0 1 1 0 1

a A 0 indicates that the corresponding signal is off. A 1 indicates that the corresponding signal is on. b Emission intensity at 335 nm (λexc ) 275 nm). c Emission intensity at 401 nm (λexc ) 357 nm). d Emission intensity at 544 nm (λexc ) 441 nm).

AN channel through one AND, one OR, and two NOT operators. Only the two inputs I1 and I3 control the output O1 of the TE channel through one NOT and one OR gate. The input I2 has no influence on the output of this channel (Table 1). Similarly, the output of the NA channel is not affected by any of the three inputs. This particular channel is always off and O1 is always equal to 0. Experimental Section The spiropyran derivative SP was synthesized following a published procedure.15 The fluorescent probes NA, AN, and TE were purchased from Aldrich and used as received. MeCN was purchased from EM Science and distilled over CaH2 under N2. The optical network shown in Figure 2 was assembled in the sample compartment of an emission spectrometer (Varian Cary Eclipse). Cell B was irradiated for 5 min at 254 nm (Mineralight UVGL-25 lamp) or for 15 min at 524 nm (ColeParmer Fiber Optic Illuminator 9745-00). When the simultaneous irradiation of cell B with ultraviolet and visible light was required, the same light sources and wavelengths were employed with an irradiation time of 15 min. The emission intensity was measured immediately after, rather than during, the application of the light inputs.

Conclusions Figure 5. The combinational logic circuits equivalent to the AN and TE channels convert the input I1, I2, and I3 into the output O1 through AND, NOT, and OR operations.

off (a-j in Figure 4 top) and the eight input strings are all transduced into a 0. In the case of the AN channel, the output O1 is on (a in Figure 4 center) when the molecular switch is state SP. Only the three input strings 000, 010, and 011 satisfy these conditions and are transduced into a 1. The other five input strings are all converted into a 0. In the case of the TE channel, the output O1 is on (a and i in Figure 4 bottom) when the molecular switch is in state SP or MEH. The six input strings 000, 001, 010, 011, 101, and 111 satisfy these conditions and are transduced into a 1. The other two input strings are converted into a 0. The combinational logic circuits5 equivalent to the transduction protocols of the AN and TE channels are illustrated in Figure 5. The three inputs I1, I2, and I3 (red in Figures 2 and 5) modulate the output O1 (yellow in Figures 2 and 5) of the

We have demonstrated that optical signals can be communicated between complementary molecules and that complex logic operations can be performed by using molecule-based systems. Our molecular switch can modulate the intensity of a traveling optical signal in response to optical and chemical stimulations. This gating mechanism does not at all require the participation of undesired electrical signals. In addition, the wavelength dependence of the switching protocol permits the control of three independent optical channels with a single switch. This simple operating principle to gate optical signals with optical signals might lead to the development of all-optical switches based on molecular components. First, however, the switching mechanism, timescale, and fatigue resistance must be assessed and methods to avoid the unpractical use of chemical inputs and volatile organic solvents must be identified. Acknowledgment. We thank the University of Miami for

financial support. JA016920E

J. AM. CHEM. SOC.

9

VOL. 124, NO. 9, 2002 2007